吉首大学学报(自然科学版)

• 计算机 • 上一篇    下一篇

一种分区BP人工神经网络图像差值算法

钱育蓉,于炯,英昌甜,杨兴耀   

  1. (新疆大学软件学院,新疆 乌鲁木齐 830008)
  • 出版日期:2016-05-25 发布日期:2016-06-24
  • 作者简介:钱育蓉(1980—),女(满族),山东德州人,新疆大学软件学院副教授,博士,主要从事图像处理、网络计算研究;于炯(1964—),男,新疆乌鲁木齐人,新疆大学软件学院教授,博士生导师,主要从事网络计算研究.
  • 基金资助:

    国家自然科学基金资助项目(61462079,61363083,61262088);新疆维吾尔自治区青年博士科技人才创新项目(2013731004);新疆维吾尔自治区高校科研计划项目(XJEDU2012I10);新疆维吾尔自治区自然科学基金资助项目(2013211A011,2011211A011)

Regional Image Interpolation Algorithms with Back-Propagation Artificial Neural Network

QIAN Yurong,YU Jiong,YING Changtian,YANG Xingyao   

  1. (School of Software,Xinjiang University,Urumqi 830008,China)
  • Online:2016-05-25 Published:2016-06-24

摘要:

为了提高图像插值质量,利用前向反馈人工神经网络(BP-ANN)的自学习、自适应和泛化能力,开展分区BP-ANN图像差值研究.将图像中待插值像素划分为光滑区和边缘区,每个区分别对应1个BP-ANN进行图像差值操作,并通过3组实验确定分区BP-ANN的网络结构、采点模式和插值流程.结果表明,采用8-16-1拓扑结构的BP-ANN算法可达到图像的可视化质量和时间之间的最佳平衡点;与经典线性均值(LA)图像插值算法相比,分区BP-ANN算法在保持最佳视觉效果的前提下,峰值性噪比高约0.593 8 dB.

关键词: 插值, 前向反馈人工神经网络, 线性均值插值, 峰值性噪比, 网络拓扑

Abstract:

In order to improve the quality of image interpolation,back-propagation artificial neural network (BP-ANN) with self-learning,adaptive and generalization ability has been used to carry out the regional image interpolation research.Missing pixel in image is divided into smooth region and edge region,each region corresponding to a BP-ANN interpolation operation.To identify the topology structure,sampling mode and the interpolation process of regional BP-ANN,three experiments were performed.The experimental results show that best balance between CPU-time consumption and visual quality lies in 8-16-1 topology of BP-ANN;compared with the classical LA interpolation algorithm,our proposed algorithm provides 0.593 8 dB higher PSNR while performed better visual quality.

Key words: interpolation, back propagation artificial neural network, linear average interpolation, peak signal to noise radio, network topology

公众号 电子书橱 超星期刊 手机浏览 在线QQ